Simulation Advices

Below are a few basic advices newbies unfortunately often skip.

1. Validation of your physics

This will sound obvious but the physical models you're using need to be validated.
Are they really adapted to you problem (energy range, accuracy, ...)?
So the first thing to do is to make sure you're using a good model.
Several approaches can be used:

1. Compare your model with web data using a basic geometry (particle X impinging on a slab of material Y at all energies)

• photons: NIST XCOM (myXCOM.tgz)*
• protons: NIST PSTAR or SRIM (mySRIM_proton.tgz)*
• alpha: NIST ASTAR or SRIM (mySRIM_alpha.tgz)*
• electrons: NIST ESTAR (myESTAR.tgz)*
• LCPs (Z>2): SRIM
• neutrons: Janis + Watch out the kinematics! See e.g. this exemplary document (bug tracking and fixing).
• For nuclear data, please see here
• For radiation protection purposes, you could try to recover the values computed in this book (available at RAMSES).

*NB: The archives provided contain my own reformatted data for all elements (no compounds). They should not contain any errors. If you find one, please let me know.
To plot them quickly, you can type this one-line command in the Root prompt which will display the 4th column versus the 1st (energy). The header will be skipped properly.

----------------------------------------------------------------------------
TGraph *graph = new TGraph ("NIST_ESTAR_08.dat","%lg %*s %*s %lg") ; graph->Draw("APL") ; gPad->SetLogx() ; gPad->SetLogy() ;
----------------------------------------------------------------------------

2. Simulate a complete benchmark experiment (experimental data everybody agrees upon)

• See e.g. the NEA benchmarks page

Furthermore, many articles on code validation can be found in the literature (NIM, IEEE-TNS, ...).
So check them out!

2. Systematical uncertainties

Another important aspect to take into account is systematical uncertainties.
A nice start is provided in the 2 following workshops which consists of a set of 8 different problems:

• EURADOS/CONRAD 2003
  
  • problems: here & here
    
  • problems + solutions: here
    
• EURADOS/CONRAD 2007
  
  • problems: here
  • solutions: here >> See "Summary of Pn" slides
  • problems + solutions: [xxx:/home/mtrocme/public/mcnp/docs_ramses/workshop/CONRAD_07.pdf]

The emphasis is put on geometry, densities, cross-sections...
It contains loads of tips! So please, give it a try!